Examining the stability of SAEBRS scores, classifications, and latent profiles

We investigated the stability of the Social, Academic, and Emotional Behavior Risk Screener (SAEBRS) using a naturalistic sample of teacher-report ratings for K-5 students (N = 1253) across three universal screening occasions (i.e., fall, winter, and spring) within one school year. We analyzed the stability of SAEBRS raw scores and classifications derived from these scores between screening occasions. Results showed strong raw score stability and moderate-to-strong classification stability for the SAEBRS total scale and subscales, with the Emotional Behavior subscale demonstrating the weakest relative stability. We also explored latent profiles indicated by SAEBRS subscale scores, the stability of membership in these profiles across occasions, and the role of student sociodemographic characteristics in predicting membership within these profiles. Findings indicated that two general risk profiles (i.e., at-risk and not at-risk) were identified and that membership in these profiles had strong stability across screening occasions. Moreover, we found that students' grade level, sex, race/ethnicity, special education, and free/reduced-price lunch status significantly predicted membership in the at-risk profile. Overall, results suggest that SAEBRS raw scores, classifications, and latent profiles all demonstrate good stability across three screening occasions within one school year. We discuss implications of these findings related to using the SAEBRS for universal screening in elementary settings and highlight key directions for future research.